SieveStack uses Nebius and TractoAI infrastructure to generate high-fidelity synthetic data through molecular dynamics simulations and train a multi-layer stack of foundational models on the resulting data. The work supports research into interactions between drugs and biological targets. Its goal is to unlock treatment pathways for hard-to-treat diseases. It combines physics-based molecular modeling with machine learning to simulate molecular interactions and predict laboratory experiment results. SieveStack’s own site and public posts describe the collaboration. Nebius describes SieveStack as building the world’s largest molecular simulation dataset.
Nebius’s customer story describes TractoAI’s unified compute platform, orchestration and parallelization that maintain GPU utilization above 90%. Training uses FP32 for accuracy and BF16 for performance. The story reports 30–50% shorter delays between test runs and production and training accelerated by up to four times. Nebius’s life-sciences page also reports 50% lower memory use and drug-target interactions revealed beyond the reach of laboratory experiments.
SieveStack worked with Nebius and TractoAI support teams to prototype, debug and scale its models. SieveStack says these services let the team focus on science instead of infrastructure. A later Nebius post describes reliable GPU access through Nebius as enabling the team to focus on discovery.
Nebius describes SieveStack using Nebius and TractoAI infrastructure to generate synthetic molecular simulation data and train a multi-layer stack of foundational models. It reports TractoAI-driven orchestration, GPU utilization above 90%, 30–50% shorter delays between test runs and production and training accelerated by up to four times.
Nebius Customer story24 Sept 2026 · Nebius · Web pageNebius describes SieveStack using Nebius and TractoAI infrastructure to generate synthetic molecular simulation data and train a multi-layer stack of foundational models. It reports TractoAI-driven orchestration, GPU utilization above 90%, 30–50% shorter delays between test runs and production and training accelerated by up to four times.
SieveStack company account
28 Jul 2025 · LinkedIn · SieveStack company account · Web page
SieveStack says Nebius and TractoAI let it focus on science while building scalable drug-discovery models. In the reshared post, Nebius names SieveStack as a customer generating molecular dynamics simulations and training foundational models using FP32 and BF16. Nebius reports GPU utilization above 90% and support from Nebius and TractoAI teams to prototype, debug and scale the models.
SieveStack company account28 Jul 2025 · LinkedIn · SieveStack company account · Web pageSieveStack says Nebius and TractoAI let it focus on science while building scalable drug-discovery models. In the reshared post, Nebius names SieveStack as a customer generating molecular dynamics simulations and training foundational models using FP32 and BF16. Nebius reports GPU utilization above 90% and support from Nebius and TractoAI teams to prototype, debug and scale the models.
Nebius Describes SieveStack using Nebius GPU access for drug-discovery model research
3 Mar 2026 · Nebius · Web page
Nebius describes SieveStack combining physics-based molecular modeling with machine learning to simulate molecular interactions and predict laboratory results. The post says reliable GPU access through Nebius lets SieveStack focus on discovery.
Nebius Describes SieveStack using Nebius GPU access for drug-discovery model research3 Mar 2026 · Nebius · Web pageNebius describes SieveStack combining physics-based molecular modeling with machine learning to simulate molecular interactions and predict laboratory results. The post says reliable GPU access through Nebius lets SieveStack focus on discovery.
Cloud for Scientific AI and Healthcare
8 Oct 2026 · nebius.com · Web page
Nebius describes SieveStack generating molecular dynamics simulations and training foundational models with FP32 and BF16 precision. It reports Nebius and TractoAI support for model development, GPU utilization above 90%, training accelerated by two to four times and 50% lower memory use.
Cloud for Scientific AI and Healthcare8 Oct 2026 · nebius.com · Web pageNebius describes SieveStack generating molecular dynamics simulations and training foundational models with FP32 and BF16 precision. It reports Nebius and TractoAI support for model development, GPU utilization above 90%, training accelerated by two to four times and 50% lower memory use.
SieveStack develops molecular simulation tools and AI models for researchers working on drug discovery. It combines physics-based modeling, biochemistry and molecular dynamics data to study how compounds interact with disease targets over time. Its products include BALEEN, a ligand-based AI model that learns binding-related features from molecular structures and is available by request, and a covalent modeling tool offered through Superbio.ai. The latter predicts binding poses using flexible protein side chains, simulations and automated scoring.
SieveStack uses Nebius and TractoAI infrastructure to generate high-fidelity synthetic data through molecular dynamics simulations and train a multi-layer stack of foundational models on the resulting data. The work supports research into interactions between drugs and biological targets. Its goal is to unlock treatment pathways for hard-to-treat diseases. It combines physics-based molecular modeling with machine learning to simulate molecular interactions and predict laboratory experiment results. SieveStack’s own site and public posts describe the collaboration. Nebius describes SieveStack as building the world’s largest molecular simulation dataset.
Nebius’s customer story describes TractoAI’s unified compute platform, orchestration and parallelization that maintain GPU utilization above 90%. Training uses FP32 for accuracy and BF16 for performance. The story reports 30–50% shorter delays between test runs and production and training accelerated by up to four times. Nebius’s life-sciences page also reports 50% lower memory use and drug-target interactions revealed beyond the reach of laboratory experiments.
SieveStack worked with Nebius and TractoAI support teams to prototype, debug and scale its models. SieveStack says these services let the team focus on science instead of infrastructure. A later Nebius post describes reliable GPU access through Nebius as enabling the team to focus on discovery.
Nebius describes SieveStack using Nebius and TractoAI infrastructure to generate synthetic molecular simulation data and train a multi-layer stack of foundational models. It reports TractoAI-driven orchestration, GPU utilization above 90%, 30–50% shorter delays between test runs and production and training accelerated by up to four times.
Nebius Customer story24 Sept 2026 · Nebius · Web pageNebius describes SieveStack using Nebius and TractoAI infrastructure to generate synthetic molecular simulation data and train a multi-layer stack of foundational models. It reports TractoAI-driven orchestration, GPU utilization above 90%, 30–50% shorter delays between test runs and production and training accelerated by up to four times.
SieveStack company account
28 Jul 2025 · LinkedIn · SieveStack company account · Web page
SieveStack says Nebius and TractoAI let it focus on science while building scalable drug-discovery models. In the reshared post, Nebius names SieveStack as a customer generating molecular dynamics simulations and training foundational models using FP32 and BF16. Nebius reports GPU utilization above 90% and support from Nebius and TractoAI teams to prototype, debug and scale the models.
SieveStack company account28 Jul 2025 · LinkedIn · SieveStack company account · Web pageSieveStack says Nebius and TractoAI let it focus on science while building scalable drug-discovery models. In the reshared post, Nebius names SieveStack as a customer generating molecular dynamics simulations and training foundational models using FP32 and BF16. Nebius reports GPU utilization above 90% and support from Nebius and TractoAI teams to prototype, debug and scale the models.
Nebius Describes SieveStack using Nebius GPU access for drug-discovery model research
3 Mar 2026 · Nebius · Web page
Nebius describes SieveStack combining physics-based molecular modeling with machine learning to simulate molecular interactions and predict laboratory results. The post says reliable GPU access through Nebius lets SieveStack focus on discovery.
Nebius Describes SieveStack using Nebius GPU access for drug-discovery model research3 Mar 2026 · Nebius · Web pageNebius describes SieveStack combining physics-based molecular modeling with machine learning to simulate molecular interactions and predict laboratory results. The post says reliable GPU access through Nebius lets SieveStack focus on discovery.
Cloud for Scientific AI and Healthcare
8 Oct 2026 · nebius.com · Web page
Nebius describes SieveStack generating molecular dynamics simulations and training foundational models with FP32 and BF16 precision. It reports Nebius and TractoAI support for model development, GPU utilization above 90%, training accelerated by two to four times and 50% lower memory use.
Cloud for Scientific AI and Healthcare8 Oct 2026 · nebius.com · Web pageNebius describes SieveStack generating molecular dynamics simulations and training foundational models with FP32 and BF16 precision. It reports Nebius and TractoAI support for model development, GPU utilization above 90%, training accelerated by two to four times and 50% lower memory use.
SieveStack develops molecular simulation tools and AI models for researchers working on drug discovery. It combines physics-based modeling, biochemistry and molecular dynamics data to study how compounds interact with disease targets over time. Its products include BALEEN, a ligand-based AI model that learns binding-related features from molecular structures and is available by request, and a covalent modeling tool offered through Superbio.ai. The latter predicts binding poses using flexible protein side chains, simulations and automated scoring.
